Modified Use of e-cigarettes and Marketing on YouTube

This project is funded by a 3-year National Institutes of Health R01 grant. Youth use social media such as YouTube to learn about e-cigarette use and are exposed to e-cigarette marketing through these means. YouTube is the world’s largest social media platform for sharing video content with over a billion users, and it can be studied to learn about trends in various health and risk behaviors, including e-cigarette use. This project uses YouTube as an important window into the youth appeal of e-cigarettes in the context of evolving innovations in product development and marketing strategies. CML is using cutting edge machine learning and data mining methods to identify this evidence from YouTube videos.The overall aim of this study is to identify popular ways in which youth modify e-cigarettes and their motives for doing so using focus group methods and social media informatics. Graduate and undergraduate students in CML are working on this project.


Designing Responsible AI Technologies to Curb Disinformation

This project is funded by a University of Texas at Austin six year Good Systems grant  which involves faculty from the School of Journalism and Media, School of Information, Computer Science,  Linguistics, McCombs School of Business. The project seeks to: 1) Understand Disinformation and Misinformation will advance scientific understanding of the nature of the problem itself; 2) Develop Effective Algorithmic Approaches will develop mathematical and computational approaches to modeling and mitigation; and 3) Design Effective Human-Centered Interventions will focus on human-centered interventions to enhance human sense-making skills in assessing information veracity, recognizing disinformation, and disrupting its belief, spread and re-circulation. Graduate, undergraduate, and high school students in CML are working on this project.


Social Media, Acculturation and E-cigarette Use among Mexican American College Students in South Texas

This project is funded by a 5-year National Institutes of Health R01 grant. Hispanics are particularly likely to be targeted through social media, where electronic nicotine delivery systems (ENDS) are increasingly and effectively promoted. ENDS marketing via social media increases risk for ENDS use among Hispanic young adults, including college students. In Texas, college enrollment among Hispanic high school graduates, most of whom are Mexican American, increased from 38% to 46% in 2000-2017. The primary goals of this project are a) to identify mechanisms underlying ENDS-related social media exposure and engagement, and ENDS use among Mexican American college students and b) to examine the role of acculturation as a moderator of these mechanisms. Changes in acculturation may increase substance use risk, including ENDS use. Social media are important agents of change that drive acculturation processes, which, together influence behaviors. Yet, nuanced interactions between acculturation processes experienced by Mexican Americans and social media use, on subsequent ENDS use, are understudied. Graduate and undergraduate students in CML are working on this project.


Venmo: Understanding Mobile Payments as Social Media
Payment infrastructures are going through rapid change with the rise of next generation mobile networks and smartphone ownership. From mobile wallets to rideshare apps, social payments allow users to split receipts with friends, charge exes for breakup expenses, or troll celebrities. New layers of data, sociality, and markets are being created and influenced by expanding economic imaginaries, regulations, and business models leveraging these new infrastructures. In this project we discuss how mobile payment systems have become social media. In this project, Drs. Murthy and Acker examine Venmo, a social payments platform that allows users to broadcast transactions to a social activity stream or public transaction feed. Our findings detail how transaction feeds of mobile payments support social practices, communication, and commerce with mobile devices and wireless networks. Our case study on Venmo helps develop some implications for the design, study, and impact of mobile payment infrastructures as social media.


CoVerifi: A COVID-19 News Verification System

This project seeks to make a timely intervention to the information landscape through a COVID-19 fake news, misinformation, and disinformation website. The website asks visitors to participate in crowd-based citizen science by quickly asking their opinion of whether something is fake news, misinformation, or disinformation. CoVerifi is a web application creating a news feed which combines both the power of machine learning and the power of human feedback to assess the credibility of news. Through allowing users the ability to “vote” on news content, the CoVerifi platform will allow us to release labelled data as open source, which will enable further research on preventing the spread of COVID-19-related misinformation. CoVerifi also has potential utility of being deployed at scale for combating the COVID-19 “infodemic”.


Rural Loss Estimates of Hurricane Florence Enabled by Citizen Scientists
This project was funded by Award#1902460 of the National Science Foundation’s Division Of Civil, Mechanical and Manufacturing Innovation (CMMI) . This project examines the ways in which everyday citizen scientists might be able to effectively collect perishable research-grade damage data following a hurricane that affects rural locales. This research uses Hurricane Florence in 2018 as a test case. The project uniquely leverages networks of journalists in rural areas to actively recruit local citizen scientists to document damage using multiple channels. This project directly enhances the understanding of rural damage from Florence and will impact the immediate recovery of these often marginalized and neglected communities. Furthermore, this research measures and tests the efficacy of citizen-produced versus engineer-collected damage data in rural contexts, validated by ground truth reconnaissance data. Dr. Murthy is PI on this project and continues to work with data from the project.